• New economic theories prompt us to ask a host of questions related to public policy: Does the fact that buyers and sellers possess different, “asymmetric” information about a transaction mean the transaction is “lopsided” and in need of government redress for the less-informed party? When a market leader establishes a universal standard in its industry, does it no longer need to keep improving to maintain its market leadership?
  • The new Independent Institute book Market Failure or Success draws upon cutting-edge economic research to help answer the above questions and others. Together, this research sheds light on the age-old, but always relevant question, do markets work well on their own, or do they require government intervention to correct widespread “market failure”?
  • Theories of market failure have long been based on the problem of free-riders and externalities. In recent years new theories of market failure have been developed and added to this corpus, usually drawing on newer research topics such as information economics or network effects. These new theories have been cited to help justify a revival of antitrust activism and government regulation in general.
  • Just as the older theories of market failure drew useful criticism, so the new market-failure theories are now facing rigorous criticism, although the new challenges are not well known. Market Failure or Success presents the most important articles of the new market-failure theories, along with the even newer theoretical and empirical rebuttals.
  • Challenges to the new market-failure theories share common threads. They often argue that the new market-failure theory unfairly compares economic institutions against an unobtainable ideal (the “nirvana approach”) instead of against a real-world alternative. They also argue that although new market-failure theories may have theoretical appeal, their proponents have not attempted to validate them empirically.
  • The market for used trucks, for decades the standard example of a market failure, shows no sign of problems due to adverse selection. Market forces allow consumers to be better informed than suggested by urban myths.
  • Public goods and other collective action problems can and are overcome by voluntary action, in part because cooperation is much more common than in standard economic models. Lighthouses and other textbook examples of public goods, for example, have at many times been provided apart from the government.
  • Life insurance markets are robust and show no signs of failure due to asymmetric information. In fact, prices in life insurance markets show positive selection not adverse selection.
  • The superiority of Beta to VHS video recording is largely a myth. The market correctly rewarded the superior product.
  • Consumer-protection laws and regulations are often rationalized on the grounds that without them the public would make uninformed or foolish choices. But the risks arising from imperfect knowledge prompt consumers and entrepreneurs to discover new ways to assure quality and safety, thus undercutting the case for government regulation as a “solution.”


Throughout the twentieth century, most arguments for the existence of market failure were based on theories of public goods and externalities. Recent years, however, have seen the development of new theories based on asymmetric information and network effects. In Market Failure or Success: The New Debate, economists Tyler Cowen and Eric Crampton bring together the strongest and most influential of the recent market-failure theories, as well as the evidence that markets may work better than the theories predict.

New Market Failure Theories

On first glance, the critics of markets make a strong case for market failure, as Cowen and Crampton explain in their introductory chapter. Markets are characterized by asymmetries in information between buyers and sellers. Employers may pay more than market clearing wages to encourage a more productive work force. And some decisions, once made, are difficult to reverse and may prove to have been worse than other alternatives. These problems are well articulated in influential papers by Joseph Stiglitz, George Akerlof and Paul David, usefully included as chapters 2 to 5 in this volume.

Theoretical Responses

However, even if we were to accept every market-failure argument, solutions to the alleged problems are hard to come by. Harold Demsetz reminds us in “Information and Efficiency: Another Viewpoint” (chap. 6) that simply pointing out how markets fail to reach a theoretical ideal does not condemn markets. Demsetz recommends comparative institutional analysis. It is more helpful, Demsetz explains, to show how a real-world policy or market institution compares to a realizable alternative than to compare it to a theoretical ideal that cannot be realized (a commonly practiced method Demsetz calls the “nirvana approach”). Simply put, Demsetz argues that the market-failure theorists have failed to provide reasons for us to believe that government intervention could improve outcomes.

Credit markets, for example, may be distorted by adverse selection. (For example, borrowers know their probability of repayment better than do lenders.) However, in “Do Information Frictions Justify Federal Credit Programs?” (chap. 8) Stephen D. Williamson shows that because governments are no better at verifying creditworthiness than are banks, government provision of credit will not make borrowers and lenders better off.

But we shouldn’t be so quick to cede ground to market failure theories. Even the theoretical case for numerous market failures is found wanting.

H. Lorne Carmichael argues in “Efficiency Wage Models of Unemployment: One View” (chap. 7) that efficiency wage models explain much less about the world than is frequently claimed. In addition to having shaky theoretical underpinnings, these models lead to predictions that do not correspond with real world patterns in unemployment. For example, efficiency wage models do not predict wage rigidity or cyclical unemployment.

Asymmetric information in markets is a problem. Buyers and sellers cannot always trust one another, nor are they always completely honest with each other. But this problem creates an opportunity for entrepreneurial activity, as Daniel B. Klein explains in “The Demand for and Supply of Assurance” (chap. 9). Many businesses exist solely for the purpose of bringing together buyers and sellers and thus turning isolated dealings into repeated interactions over time, thereby fostering trust. These businesses provide assurance—seals of approval recognized by buyers who would otherwise have a difficult time independently assessing quality. Where market-failure theories predict massive deadweight losses from trades that fail to occur, Klein shows that alert entrepreneurs see these losses as potential profits to be made by removing the impediments to trade.

Empirical and Experimental Responses

Empirical and experimental analyses also suggest that markets are more robust than would be predicted by the failure theories.

Paul David argues that network markets will be characterized by lock-in and path dependence. In network markets, David argues that being the first company to develop a strong user base is critically important in determining future successes. In “Beta, Macintosh and Other Fables” (chap. 10), Stan J. Liebowitz and Stephen E. Margolis show that the record does not support David’s theory. Where David takes QWERTY (the traditional English-language typewriter keyboard arrangement) as the archetypical example of inefficient path-dependent lock-in, Liebowitz and Margolis show that the case against QWERTY is ephemeral at best. Network markets need not lead to market failures. Not only do products only achieve market dominance once they have proven their superiority over alternative products, competition between standards often remains quite strong.

Adding to Stephen D. Williamson’s theoretical reasons to be skeptical of government intervention in credit markets, Allen N. Berger and Gregory F. Udell test the significance of the credit rationing theory directly and find that the data do not support the theory (chap. 11, “Some Evidence on the Empirical Significance of Credit Rationing”). They show that credit rationing “is not a significant macroeconomic phenomena.” Similarly, John Cawley and Tomas Philipson find that life insurance markets do not reflect the failures predicted by the asymmetric information critique of markets (chap. 12, “An Emprical Examination of Barriers To Trade in Insurance”). Eric W. Bond finds that the market for used vehicles, the archetype used by Akerlof in his initial “lemons” model, exhibits none of the features of the “lemons” market; used trucks sold by their owners are of no worse quality than those kept by their owners (chap. 13, “A Direct Test of the ‘Lemons’ Model: the Market for Used Pickup Trucks”).

Experimental economics provides us with additional insights into market behavior. The under-provision of public goods is a classic example of market failure. Another example offered is that people will cheat on each other whenever doing so will increase their earnings. These examples assume that the potential for free-riding on the efforts of others will routinely undermine economic cooperation. Experimental economics puts these theories to the test and finds that people do not behave in these ways.

Elizabeth Hoffman surveys the experimental literature and shows that free-riding is far from the norm, and cooperation is surprisingly common (chap. 14, “Public Choice Experiments”). Gordon Tullock finds that when individuals in a prisoner’s dilemma game are allowed to communicate with each other and choose their own partners, cooperation results (chap. 15, “Non-prisoner’s Dilemma”). R. Mark Isaac, James M. Walker and Arlington W. Williams show that, contrary to expectations, people frequently do contribute to the provision of public goods, even in large groups where we should most expect free-riding (chap. 16, “Group Size and the Voluntary Provision of Public Goods: Experimental Evidence Utilizing Large Groups”). And James Andreoni shows that kindness and cooperation are common even in rather artificial experimental settings (chap. 17, “Cooperation in Public-Goods Experiments: Kindness or Confusion”).

Together with the insightful introduction by editors Tyler Cowen and Eric Crampton, Market Failure or Success provides a compelling argument for the success of markets. The world is far from perfect. But the imperfections are smaller than we might expect. And, for all their criticisms of markets, market-failure theorists have not at all provided a compelling case against government failure, that government could work at all better.

About the Editors

Tyler Cowen is general director of the Mercatus Center and the James M. Buchanan Center for Political Economy at George Mason University.

Eric Crampton is a graduate fellow at the Center for the Study of Public Choice, George Mason University.

Product Details

$95.00 (hardcover). 367 pages. 6 x 9 inches. ISBN 978-1-84376-025-2.
$35.00 (softcover). 367 pages. 6 x 9 inches. ISBN 978-1-84376-085-6.